Kedis ku xidhan hagaha · Adag Heerka

Kediska Is-Distille DINO

Dib u eegis cusboonaysiinta macalinka DINO, joogteynta dalagga, ka hortagga burburka, iyo habka xog-wareedka ee taageera sifooyinka DINov2 ee la wareejin karo.

Wadooyinka haga ee la xidhiidhaDino Self Distillation
Su'aal 1 ee 8

Maxay u taagan tahay 'MAYA' ee DINO?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Maxay u taagan tahay 'MAYA' ee DINO?
    • Sumad la'aan
    • Ma jiro wax hagaajin
    • Wax soo saarka neerfaha
    • Ujeedada caadiga ah
  2. Sidee loo cusboonaysiiyaa miisaanka shabakadda macalinka gudaha DINO?
    • Iyada oo si toos ah ugu soo degtay khasaaraha
    • Si aan kala sooc lahayn tallaabo kasta
    • Sida celceliska dhaqdhaqaaqa jibbaaran ee miisaanka ardayga
    • Waa laga qaboojiyey qaab horay loo tababaray
  3. Dhibaato intee le'eg ayay ka hortagtaa udub dhexaadinta iyo fiiqidda natiijada macalinka?
    • Kulaylka badan ee GPU-ga
    • Qaabku u burburo wax soo saarka joogtada ah
    • Soo dejinta xogta oo gaabis ah
    • Qallooca midabka
  4. Istaraatiijiyada dalagyada badan ee DINO, aragtidee ayuu macalinku helaa?
    • Kaliya dalagyada yaryar ee maxaliga ah
    • Xirmooyinka dhawaaqa random
    • Noocyo cirro leh oo keliya
    • Kaliya dalagyada caalamiga ah
  5. What emergent property did the original DINO study observe in vision-transformer attention maps?
    • Every map reconstructs an exact depth measurement
    • Some maps highlight object structure without segmentation-label training
    • Attention is trained directly on human-drawn masks
    • All heads produce perfect object boundaries on every image
  6. How did DINOv2 extend the visual-feature approach of the original DINO?
    • It scaled self-supervised training with curated data for transferable visual features
    • It replaced visual inputs with text-only training
    • It required segmentation labels for all pretraining images
    • It removed the student-teacher learning approach entirely
  7. In DINO training, which network receives gradients from the student-teacher cross-entropy objective?
    • Only the teacher, while the student stays frozen
    • Both networks through the same backpropagation update
    • Neither network because the outputs are fixed labels
    • The student, while teacher parameters follow a moving average
  8. What does using DINOv2 as a frozen backbone mean for a downstream task?
    • It must be retrained from scratch for every input image
    • No task-specific output component can be attached
    • Its pretrained feature-extractor weights remain fixed while task-specific components use its features
    • The pretrained features can only reproduce the original training images